ITMO_FS API¶
This is the full API documentation of the ITMO_FS toolbox.
ITMO_FS.filters: Filter methods¶
ITMO_FS.filters.univariate: Univariate filter methods¶
filters.univariate.VDM([weighted]) |
Creates Value Difference Metric builder http://aura.abdn.ac.uk/bitstream/handle/2164/10951/payne_ecai_98.pdf?sequence=1 https://www.jair.org/index.php/jair/article/view/10182 |
filters.univariate.UnivariateFilter(measure) |
Basic interface for using univariate measures for feature selection. |
Measures for univariate filters¶
filters.univariate.fit_criterion_measure(X, y) |
|
filters.univariate.f_ratio_measure(X, y) |
Calculates Fisher score for features. |
filters.univariate.gini_index(X, y) |
Gini index is a measure of statistical dispersion. |
filters.univariate.su_measure(X, y) |
SU is a correlation measure between the features and the class calculated, via formula SU(X,Y) = 2 * I(X|Y) / (H(X) + H(Y)) |
filters.univariate.spearman_corr(X, y) |
Calculates spearman correlation for each feature. |
filters.univariate.pearson_corr(X, y) |
Calculates pearson correlation for each feature. |
filters.univariate.fechner_corr(X, y) |
Calculates Sample sign correlation (Fechner correlation) for each feature. |
filters.univariate.kendall_corr(X, y) |
Calculates Sample sign correlation (Kendall correlation) for each feature. |
filters.univariate.reliefF_measure(X, y[, …]) |
Counts ReliefF measure for each feature |
filters.univariate.chi2_measure(X, y) |
Calculates score for the test chi-squared statistic from X. |
filters.univariate.information_gain(X, y) |
Calculates mutual information for each feature by formula, I(X,Y) = H(X) - H(X|Y) |
Cutting rules for univariate filters¶
ITMO_FS.filters.multivariate: Multivariate filter methods¶
filters.multivariate.DISRWithMassive([…]) |
Creates DISR (Double Input Symmetric Relevance) feature selection filter based on kASSI criterin for feature selection which aims at maximizing the mutual information avoiding, meanwhile, large multivariate density estimation. |
filters.multivariate.FCBFDiscreteFilter() |
Creates FCBF (Fast Correlation Based filter) feature selection filter based on mutual information criteria for data with discrete features This filter finds best set of features by searching for a feature, which provides the most information about classification problem on given dataset at each step and then eliminating features which are less relevant than redundant |
filters.multivariate.MultivariateFilter(…) |
Provides basic functionality for multivariate filters. |
filters.multivariate.STIR([n_features_to_keep]) |
Feature selection using STIR algorithm. |
filters.multivariate.TraceRatioFisher(…) |
Creates TraceRatio(similarity based) feature selection filter performed in supervised way, i.e fisher version |
filters.multivariate.MIMAGA(mim_size, …) |
Measures for multivariate filters¶
filters.multivariate.MIM(selected_features, …) |
Mutual Information Maximization feature scoring criterion. |
filters.multivariate.MRMR(selected_features, …) |
Minimum-Redundancy Maximum-Relevance feature scoring criterion. |
filters.multivariate.JMI(selected_features, …) |
Joint Mutual Information feature scoring criterion. |
filters.multivariate.CIFE(selected_features, …) |
Conditional Infomax Feature Extraction feature scoring criterion. |
filters.multivariate.MIFS(selected_features, …) |
Mutual Information Feature Selection feature scoring criterion. |
filters.multivariate.CMIM(selected_features, …) |
Conditional Mutual Info Maximisation feature scoring criterion. |
filters.multivariate.ICAP(selected_features, …) |
Interaction Capping feature scoring criterion. |
filters.multivariate.DCSF(selected_features, …) |
Dynamic change of selected feature with the class scoring criterion. |
filters.multivariate.CFR(selected_features, …) |
The criterion of CFR maximizes the correlation and minimizes the redundancy. |
filters.multivariate.MRI(selected_features, …) |
Max-Relevance and Max-Independence feature scoring criteria. |
filters.multivariate.IWFS(selected_features, …) |
Interaction Weight base feature scoring criteria. |
filters.multivariate.generalizedCriteria(…) |
This feature scoring criteria is a linear combination of all relevance, redundancy, conditional dependency Given set of already selected features and set of remaining features on dataset X with labels y selects next feature. |
ITMO_FS.filters.unsupervised: Unsupervised filter methods¶
filters.unsupervised.TraceRatioLaplacian(…) |
Creates TraceRatio(similarity based) feature selection filter performed in unsupervised way, i.e laplacian version |
ITMO_FS.filters.sparse: Sparse filter methods¶
filters.sparse.MCFS(d[, k, p, scheme, sigma]) |
Performs the Unsupervised Feature Selection for Multi-Cluster Data algorithm. |
filters.sparse.NDFS(p[, c, k, alpha, beta, …]) |
Performs the Nonnegative Discriminative Feature Selection algorithm. |
filters.sparse.RFS(p[, gamma, …]) |
Performs the Robust Feature Selection via Joint L2,1-Norms Minimization algorithm. |
filters.sparse.SPEC(p[, k, gamma, sigma, …]) |
Performs the Spectral Feature Selection algorithm. |
filters.sparse.UDFS(p[, c, k, gamma, l, …]) |
Performs the Unsupervised Discriminative Feature Selection algorithm. |
ITMO_FS.ensembles: Ensemble methods¶
ITMO_FS.ensembles.measure_based: Measure based ensemble methods¶
ensembles.measure_based.WeightBased(filters) |
ITMO_FS.ensembles.model_based: Model based ensemble methods¶
ensembles.model_based.BestSum(models, …) |
ITMO_FS.ensembles.ranking_based: Ranking based ensemble methods¶
ensembles.ranking_based.Mixed(filters) |
Performs feature selection based on several filters, selecting features this way: Get ranks from every filter from input. |
ITMO_FS.embedded: Embedded methods¶
embedded.MOS([model, loss, seed]) |
Performs Minimizing Overlapping Selection under SMOTE (MOSS) or under No-Sampling (MOSNS) algorithm. |
ITMO_FS.hybrid: Hybrid methods¶
hybrid.FilterWrapperHybrid(filter_, wrapper) |
|
hybrid.Melif(filter_ensemble[, scorer, verbose]) |
ITMO_FS.wrappers: Wrapper methods¶
ITMO_FS.wrappers.deterministic: Deterministic wrapper methods¶
wrappers.deterministic.AddDelWrapper(…[, …]) |
Creates add-del feature wrapper |
wrappers.deterministic.BackwardSelection(…) |
Backward Selection removes one feature at a time until the number of features to be removed is reached. |
wrappers.deterministic.RecursiveElimination(…) |
Performs a recursive feature elimination until the required number of features is reached. |
wrappers.deterministic.SequentialForwardSelection(…) |
Sequentially Adds Features that Maximises the Classifying function when combined with the features already used TODO add theory about this method |
Deterministic wrapper function¶
wrappers.deterministic.qpfs_wrapper(X, y, alpha) |
Performs Quadratic Programming Feature Selection algorithm. |
ITMO_FS.wrappers.randomized: Randomized wrapper methods¶
wrappers.randomized.HillClimbingWrapper(…) |
|
wrappers.randomized.SimulatedAnnealing(…) |
Performs feature selection using simulated annealing |
wrappers.randomized.TPhMGWO([wolfNumber, …]) |
Performs Grey Wolf optimization with Two-Phase Mutation |